心脏瓣膜置换术后认知功能障碍的围术期危险因素分析:一项回顾性病例对照研究
Perioperative Risk Factors for Postoperative Cognitive Dysfunction after Cardiac Valve Replacement: A Retrospective Case-Control Study
摘要: 目的:探究心脏瓣膜置换患者术后发生认知功能障碍(POCD)的围术期危险因素,并评估危险因素对POCD的预测价值。方法:本研究回顾性纳入了2024年1月至2024年12月间于青岛大学附属医院接受心脏瓣膜置换术的患者的临床资料。通过多因素逻辑回归模型探索发生术后认知功能障碍的围术期风险因素并构建联合预测因子,采用受试者工作特征曲线(ROC)和曲线下面积(AUC)分析这些风险因素对术后认知功能障碍的预测价值。结果:共208例患者被纳入分析,POCD的发生率为23.08%。多因素逻辑回归分析显示术中尿量(OR = 0.998; 95%CI: 0.997~0.999; P = 0.004)、术中丙泊酚消耗量(OR = 1.002; 95%CI: 1.001~1.003; P = 0.003)、术前C反应蛋白(OR = 0.939; 95%CI: 0.895~0.985; P = 0.01)、术后1小时的中性粒淋巴细胞比值(OR = 1.181; 95%CI: 1.094~1.276; P < 0.001)、术后第1天的C反应蛋白(OR = 1.031; 95%CI: 1.013~1.049; P < 0.001)以及术后第7天睡眠状况自评量表得分(OR = 1.161; 95%CI: 1.088~1.238; P < 0.001)是心脏瓣膜置换患者发生POCD的独立影响因素,通过逻辑回归模型构建的联合预测因子可以对POCD进行有效预测(AUC = 0.7845; OR = 1.671; 95%CI: 1.446~1.930)。结论:基于围术期风险因素构建的联合预测因子可用于对心脏瓣膜置换术后的认知功能障碍进行早期预测。
Abstract: Objective: To investigate the perioperative risk factors for the postoperative cognitive dysfunction (POCD) in patients undergoing cardiac valve replacement and to evaluate the predictive value of these risk factors for POCD. Methods: This study retrospectively included the clinical data of patients who underwent cardiac valve replacement at the Affiliated Hospital of Qingdao University from January 2024 to December 2024. A multivariate logistic regression model was used to explore the perioperative risk factors for postoperative cognitive dysfunction (POCD) and to construct a composite predictor. The predictive value of these risk factors for POCD was analyzed using the receiver operating characteristic (ROC) curve and the area under the curve (AUC). Results: A total of 208 patients were included in the analysis, with an incidence of POCD of 23.08%. Multivariate logistic regression analysis revealed that intraoperative urine output (OR = 0.998; 95% CI: 0.997~0.999; P = 0.004), intraoperative propofol consumption (OR = 1.002; 95% CI: 1.001~1.003; P = 0.003), preoperative C-reactive protein (CRP) level (OR = 0.939; 95% CI: 0.895~0.985; P = 0.01), neutrophil-to-lymphocyte ratio (NLR) at 1 hour postoperatively (OR = 1.181; 95% CI: 1.094~1.276; P < 0.001), CRP level on the first postoperative day (OR = 1.031; 95% CI: 1.013~1.049; P < 0.001), and Self-Rating Scale of Sleep (SRSS) score on the seventh postoperative day (OR = 1.161; 95% CI: 1.088~1.238; P < 0.001) were independent influencing factors for the development of POCD in patients undergoing cardiac valve replacement. The composite predictor constructed using the logistic regression model effectively predicted POCD (AUC = 0.7845; OR = 1.671; 95%CI: 1.446~1.930). Conclusion: The composite predictor constructed based on perioperative risk factors can be used for early prediction of postoperative cognitive dysfunction following cardiac valve replacement.
文章引用:李宗笑. 心脏瓣膜置换术后认知功能障碍的围术期危险因素分析:一项回顾性病例对照研究[J]. 临床医学进展, 2025, 15(4): 2370-2381. https://doi.org/10.12677/acm.2025.1541190

1. 引言

术后认知功能障碍(Postoperative cognitive dysfunction, POCD)是心脏瓣膜置换术后的常见并发症,表现为术前认知功能正常的患者,在手术后出现学习、记忆、信息加工、定向和注意力等认知功能的显著下降[1]。临床研究发现约35.7%至58%的患者会在心脏手术后1个月内出现不同程度的认知功能障碍,发生率明显高于非心脏手术患者[2]-[5],这可能与心脏手术后更严重的全身炎症反应有关。手术诱发的无菌性炎症会造成中枢神经系统神经元凋亡和突触功能失调,进而损害患者的神经认知功能[6] [7]而心脏瓣膜置换手术由于创伤较大、术中行体外循环以及缺血再灌注损伤等因素,引起的全身炎症反应更加明显,因此心脏瓣膜置换患者是术后发生认知功能障碍的高危群体[8]。发生POCD将对患者的预后造成不良影响,不仅会延长ICU住院时间并降低术后生活质量,还增加了术后远期罹患持久性认知功能障碍和痴呆的风险[9]-[11]。因此,探索心脏瓣膜置换术后发生POCD的高危因素对于改进医疗行为具有重要价值,这将有助于提前识别高危患者并指导采取相应医疗干预措施,以改善患者的认知结局。本研究旨在探讨心脏瓣膜置换患者术后早期认知功能的恢复情况,并寻找发生POCD的高危因素,为围术期认知功能保护提供临床医学证据。

2. 资料和方法

2.1. 研究对象

本研究为一项回顾性病例对照研究,得到了青岛大学附属医院伦理委员会的批准[伦理审查批件号:QYFYEC2024-24-01],收集了2024年01月至2024年12月间,所有在青岛大学附属医院接受心脏瓣膜置换(Heart Valve Replacement, HVR)手术患者的临床资料。纳入标准包括:① 接受开胸心脏瓣膜置换手术;② 年龄大于18周岁;③ 术前完成了基线认知功能测试,简易精神状态检查量表(Minimum Mental State Examination, MMSE)得分大于18。排除标准为:① 与本研究相关的临床资料不完整;② 由于术后死亡、失访、虚弱等原因导致认知功能评估资料不完整;③ 既往存在精神疾病史或精神药物使用史;④ 既往脑卒中或脑外伤史;⑤ 罹患神经系统退行性疾病;⑥ 术中死亡。共有232名接受心脏瓣膜置换手术的患者被纳入本研究,其中2例因临床资料不完整、17例因术后未完成认知功能评估、3例因患有抑郁或精神分裂症而长期服用精神药物、1例因脑卒中后遗症、1例因术中死亡而被排除,最终对208名患者的临床数据进行了统计分析。

2.2. 研究方法

2.2.1. 临床资料

临床资料包括患者的基线特征、手术资料和围术期全身炎症指标。基线特征包括性别、年龄、体质指数(bodymass index, BMI)、左心室射血分数(Left Ventricular Ejection Fraction, LVEF)、受教育时长、房颤病史、糖尿病史、查尔森合并症指数。手术资料包括手术类型(单瓣膜置换或多瓣膜置换)、手术时长、体外循环时长、术中尿量、术中失血量、术中麻醉药物消耗量(舒芬太尼、咪达唑仑、丙泊酚、右美托咪定)。全身炎症指标包括手术前1天(PTS1)、手术后1小时(POD0)、手术后1天(POD1)、手术后2天(POD2)等几个时间点血液标本中C反应蛋白(C-reactive protein, CRP)、中性粒淋巴细胞比值(neutrophil-to-lymphocyte ratio, NLR)、白细胞(white blood cell, WBC)计数等炎症标志物水平。本研究还收集了患者在术后第7天的视觉模拟评分(Visual Analogue Scale, VAS)和睡眠状况自评量表(Self-Rating Scale of Sleep, SRSS)得分。

2.2.2. 术后认知功能障碍的诊断标准

本研究使用蒙特利尔认知评估(Montreal Cognitive Assessment, MoCA)量表评估患者围术期的认知功能。从病历资料中收集患者在手术前1天和手术后1周时的MoCA量表得分,并根据Moller等人描述的方法[12],使用Z分数对POCD进行诊断。本研究同时对57名年龄和受教育程度与手术患者相似的志愿者进行了相同的MoCA量表评估,志愿者组患者的基线特征与手术患者组相似,见表1。若患者术后1周时MoCA得分较术前下降超过志愿者平均变化水平的两个标准差,则诊断为发生了POCD。

Table 1. Baseline comparison between the patient group and the volunteer group

1. 患者组与志愿者组基线比较

患者组(n = 104)

志愿者组(n = 57)

t/χ2

P

性别(男,%)

58 (55.8%)

31 (54.4%)

0.03

0.866

年龄(岁)

60.9 ± 9.4

60.2 ± 8.7

0.47

0.637

受教育时长(年)

8.2 ± 3.2

7.7 ± 3.1

0.93

0.354

基线MMSE得分

25.2 ± 3.2

25.3 ± 3.3

−0.21

0.832

基线MoCA得分

20.8 ± 4.1

21.4 ± 4.4

−0.80

0.428

MMSE:建议精神状态检查量表;MoCA:蒙特利尔认知评估量表。

2.2.3. 麻醉方案

所有患者均接受标准化的麻醉管理方案。进入手术室后接受心电图(Electrocardiogram, ECG)、脉搏血氧饱和度(Pulse Oximetry Saturation, SpO2)、无创血压、有创上肢动脉压、中心静脉压(右侧颈内静脉)、鼻咽温度、脑电双频指数(Bispectral Index, BIS)监测等监护。使用咪达唑仑(0.04 mg/kg,江苏恩华药业,中国南京)、舒芬太尼(0.3 μg/kg,宜昌仁福药业,中国宜昌)、依托咪酯(0.2 mg/kg,江苏恩华药业,中国南京)和顺式阿曲库铵(0.2 mg/kg,江苏恒瑞药业,中国南京)进行麻醉诱导。术中持续静脉输注丙泊酚(4~8 mg/kg/h)、舒芬太尼(0.2~0.3 μg/kg/h)、右美托咪定(0.5 μg/kg/h)顺式阿曲库铵(1.5~2.5 µg/kg/min)以维持足够的肌肉松弛效果和麻醉深度(BIS值40~60)。所有患者术后均进入心血管外科重症监护病房进行监护。

2.3. 统计分析

根据术后是否发生POCD将患者分为无POCD组和POCD组,使用SPSS 27.0软件对数据进行分析。首先对两组进行差异性比较,使用Shapiro-Wilk检验评估连续结果分布的正态性,正态分布计量资料描述为平均值 ± 标准差,使用独立样本t检验比较。偏态分布计量资料描述为中位数(25%分位数,75%分位数),使用Kruskal-waills检验比较。分类资料以百分比(%)表示,使用Pearson χ2检验进行比较。通过多因素逻辑回归分析术后发生POCD的独立危险因素,并拟合多个独立影响因素构建联合预测因子模型。使用受试者工作特征曲线(Receiver Operating Characteristic curve, ROC)计算原始指标和联合预测因子的ROC曲线下面积(Area Under Curve, AUC)以评价模型对术后发生POCD的预测能力。P < 0.05认为差异有统计学意义。

3. 结果

3.1. 临床资料

本研究共分析了208名患者的临床资料,其中48名患者在术后被诊断为POCD,发病率约为23.1%。组间差异性比较显示,与无POCD组患者相比,POCD组患者术中尿量更少,咪达唑仑的消耗量更低而丙泊酚的消耗量更高,术后第7天的SRSS得分更高(P < 0.05)。在全身炎症指标的比较中,POCD组患者手术前1天的CRP、NLR、WBC计数、术后1小时的CRP、NLR以及术后第1天的CRP水平更高(P < 0.05)。见表2

Table 2. Comparison of clinical data between the two patient groups

2. 两组患者临床资料比较

无POCD组(160)

POCD组(48)

t/χ2

P

性别(男)

92 (57.50%)

24 (50.00%)

0.84

0.359

年龄(岁)

60.36 ± 9.87

62.58 ± 7.54

−1.44

0.152

BMI

24.09 ± 3.42

23.73 ± 3.09

0.65

0.517

LVEF (%)

57.89 ± 6.48

58.50 ± 5.66

−0.59

0.555

受教育时长(年)

8.19 ± 3.32

8.04 ± 2.66

0.29

0.772

房颤史

46 (28.75%)

14 (29.17%)

0.00

0.955

糖尿病史

32 (20.00%)

8 (16.67%)

0.26

0.607

查尔森合并症指数

1.11 ± 1.37

1.17 ± 1.12

−0.25

0.803

单瓣膜置换

98 (61.25%)

22 (45.83%)

3.60

0.058

多瓣膜置换

62 (38.75%)

26 (54.17%)

3.60

0.058

手术时长(min)

258.31 ± 64.80

259.58 ± 69.66

−0.12

0.907

体外循环时长(min)

121.24 ± 44.96

119.83 ± 41.96

0.19

0.847

术中尿量(ml)

1052.88 ± 592.68

727.08 ± 406.59

4.34

<0.001*

术中失血量(ml)

699.38 ± 180.06

664.58 ± 177.74

1.18

0.240

舒芬太尼(μg)

196.97 ± 46.99

183.12 ± 48.59

1.78

0.077

咪达唑仑(mg)

9.98 ± 8.92

5.67 ± 5.11

4.23

<0.001*

丙泊酚(mg)

1272.80 ± 421.49

1435.81 ± 489.58

−2.26

0.025*

右美托咪定(mg)

163.81 ± 51.86

166.03 ± 53.37

−0.24

0.807

POD7 VAS得分

4.11 ± 1.32

4.42 ± 1.57

−1.34

0.182

POD7 SRSS得分

21.16 ± 7.52

31.42 ± 8.80

−7.95

<0.001*

PTS1 CRP

1.97 (0.99, 6.21)

3.23 (2.58, 6.01)

−2.83

0.005*

POD0 CRP

2.81 (1.34, 10.36)

4.26 (2.24, 17.35)

−2.12

0.034*

POD1 CRP

68.25 (51.44, 84.39)

82.20 (61.63, 132.66)

−3.50

<0.001*

POD2 CRP

243.06 (181.76, 282.98)

264.78 (243.83, 274.86)

−1.39

0.166

PTS1 NLR

1.85 (1.42, 2.91)

2.14 (1.86, 4.35)

−2.26

0.024*

POD0 NLR

10.98 (7.39, 15.32)

21.02 (12.93, 25.22)

−6.30

<0.001*

POD1 NLR

11.78 (8.51, 16.58)

13.61 (9.08, 22.31)

−1.64

0.101

POD2 NLR

10.67 (8.13, 15.71)

14.77 (8.07, 19.90)

−1.15

0.252

PST1 WBC

5.63 (4.72, 6.90)

5.47 (4.95, 7.10)

−0.61

0.545

POD0 WBC

9.99 (7.38, 12.89)

10.67 (9.49, 17.12)

−2.18

0.029*

POD1 WBC

9.84 (8.12, 12.79)

9.93 (8.78, 11.71)

−0.51

0.608

POD2 WBC

10.77 (9.17, 13.64)

9.99 (9.15, 13.62)

−0.61

0.540

PTS1:手术前1天;POD0:手术后1小时;POD1:手术后第1天;POD2:手术后第2天;POD7:手术后第7天;CRP:C反应蛋白;NLR:中性粒细胞与淋巴细胞比值;VAS:视觉模拟评分;SRSS:睡眠状况自评量表得分。*P < 0.05。

3.2. 多因素逻辑回归分析

以术后是否发生POCD为因变量,将差异性分析中有统计学意义的临床资料指标纳入多因素逻辑回归分析。分析结果显示,术中尿量(OR: 0.998; 95% CI: 0.997~0.999; P = 0.004)、丙泊酚消耗量(OR: 1.002; 95% CI: 1.001~1.003; P = 0.003)、POD0 CRP (OR: 0.939; 95% CI: 0.895~0.985; P = 0.01);POD1 CRP (OR: 1.031; 95% CI: 1.013~1.049; P < 0.001);POD0 NLR (OR: 1.181; 95% CI: 1.094~1.276; P < 0.001)、POD7 SRSS得分(OR: 1.161; 95% CI: 1.088~1.238; P < 0.001)是心脏瓣膜置换患者术后发生POCD的独立影响因素:见表3。将POD0 CRP、POD1 CRP、POD0 NLR、POD7 SRSS得分等指标分别作为协变量X1、X2、X3、X4构建联合预测因子模型,联合预测因子方程表达式为logit(p) = −8.781 − 0.026X1 + 0.022X2 + 0.128X3 + 0.152X4,见表4

Table 3. Multivariate logistic regression analysis of postoperative DNR

3. 术后发生DNR的多因素逻辑回归分析

B

S.E

Wald χ2

OR

95%CI

P

术中尿量

−0.002

0.001

8.171

0.998

0.997~0.999

0.004*

咪达唑仑

0.019

0.044

0.185

1.019

0.935~1.111

0.667

丙泊酚

0.002

0.001

8.965

1.002

1.001~1.003

0.003*

PTS1 CRP

0.014

0.029

0.223

1.014

0.958~1.072

0.636

POD0 CRP

−0.063

0.024

6.687

0.939

0.895~0.985

0.010*

POD1 CRP

0.031

0.009

11.380

1.031

1.013~1.049

0.001*

PTS1 NLR

−0.025

0.165

0.023

0.975

0.706~1.348

0.879

POD0 NLR

0.167

0.039

17.939

1.181

1.094~1.276

<0.001*

POD0 WBC

−0.046

0.058

0.622

0.955

0.853~1.070

0.430

POD7 SRSS

0.149

0.033

20.516

1.161

1.088~1.238

<0.001*

常量

−10.677

2.023

27.846

-

-

-

PTS1:手术前1天;POD0:手术后1小时;POD1:手术后第1天;POD7:手术后第7天;CRP:C反应蛋白;NLR:中性粒细胞与淋巴细胞比值;SRSS:睡眠状况自评量表得分。*P < 0.05。

Table 4. Combined predictive model

4. 联合预测因子模型

自变量

B

S.E

Wald χ2

OR

95% CI

P

POD0 CRP

−0.026

0.018

2.104

0.975

0.942~1.009

0.147

POD1 CRP

0.022

0.007

8.481

1.022

1.007~1.037

0.004

POD0 NLR

0.128

0.031

16.722

1.137

1.069~1.209

<0.001*

POD7 SRSS

0.152

0.030

25.830

1.164

1.098~1.234

<0.001*

常量

−8.781

1.240

50.144

-

-

-

3.3. 联合预测因子对心脏瓣膜置换患者发生POCD的预测价值

通过绘制ROC曲线,计算出联合预测因子及各独立影响因素诊断发生POCD的AUC、最佳截断值、灵敏度、特异度分别为:联合预测因子(0.8854、0.3227、0.7917、0.9125)、POD0 CRP (0.6010、1.795、0.9167、0.35)、POD1 CRP (0.6667、99.67、0.4167、0.8750)、POD0 NLR (0.8000、15.94、0.7083、0.7875)、POD7 SRSS得分(0.8266、28.50、0.7917、0.875),使用联合预测因子对发生POCD进行预测的效果优于单项指标,见图1

Figure 1. ROC curve for predicting POCD after cardiac valve replacement

1. 预测心脏瓣膜置换术后发生POCD的ROC曲线

3.4. 倾向性匹配得分控制混杂因素

应用倾向性得分匹配(Propensity Score Matching, PSM),减少术中尿量、术中咪达唑仑和丙泊酚用量、PTS1CRP、PTS1NLR、POD0WBC等两组间存在显著差异性的非独立危险因素对结局分析的影响,表5显示在进行了PSM处理后,两组间上述指标的差异得到了控制。对匹配后数据进行多因素逻辑回归分析结果显示,POD0 CRP、POD1 CRP、POD0 NLR、POD7 SRSS得分等指标仍然是心脏瓣膜置换患者术后发生POCD的独立影响因素,见表6

Table 5. Comparison of balance of confounding factors between the two groups before and after PSM

5. PSM前后两组间混杂因素的均衡性比较

PSM前

PSM后

无POCD (n = 160)

POCD (n = 48)

t/Z

P

SMD

无POCD (n = 104)

POCD (n = 42)

t/Z

P

SMD

术中尿量

1052.88 ± 592.68

727.08 ± 406.59

4.338

<0.001

−0.801

824.62 ± 405.73

750.00 ± 424.26

0.993

0.322

−0.176

咪达唑仑

9.98 ± 8.92

5.67 ± 5.11

4.230

<0.001

−0.845

7.42 ± 6.33

6.19 ± 5.26

1.116

0.266

−0.234

丙泊酚

1272.80 ± 421.49

1435.81 ± 489.58

−2.262

0.025

0.333

1240.02 ± 380.13

1355.83 ± 421.10

−1.615

0.108

0.275

PTS1CRP

1.97 (0.99, 6.21)

3.23 (2.58, 6.01)

−2.830

0.005

0.050

2.30 (1.06, 5.91)

3.04 (2.50, 5.66)

−1.814

0.070

0.084

PTS1NLR

1.85 (1.42, 2.91)

2.14 (1.86, 4.35)

−2.263

0.024

0.367

1.97 (1.55, 2.76)

2.14 (1.73, 2.41)

−0.984

0.325

0.141

POD0WBC

9.99 (7.38, 12.89)

10.67 (9.49, 17.12)

−2.181

0.029

0.289

9.99 (7.46, 12.35)

10.67 (9.54, 14.69)

−1.528

0.126

0.218

PTS1:手术前1天;POD0:手术后1小时。*P < 0.05。

Table 6. Multivariate logistic regression analysis of DNR after PSM

6. PSM后DNR的多因素逻辑回归分析

B

S.E

Wald χ2

OR

95%CI

P

术中尿量

−0.001

0.001

2.530

0.999

0.998~1.000

0.112

咪达唑仑

0.042

0.050

0.715

1.043

0.946~1.150

0.398

丙泊酚

0.002

0.001

8.245

1.002

1.001~1.004

0.004*

PTS1 CRP

0.033

0.055

0.350

1.033

0.927~1.151

0.554

POD0 CRP

−0.063

0.024

6.560

0.939

0.895~0.985

0.010*

POD1 CRP

0.029

0.009

9.709

1.029

1.011~1.048

0.002*

PTS1 NLR

−0.027

0.199

0.019

0.973

0.659~1.437

0.891

POD0 NLR

0.147

0.039

14.244

1.158

1.073~1.250

<0.001*

POD0 WBC

−0.047

0.065

0.526

0.954

0.839~1.084

0.468

POD7 SRSS

0.135

0.033

16.319

1.144

1.072~1.221

<0.001*

常量

−10.394

2.069

25.245

-

-

-

PTS1:手术前1天;POD0:手术后1小时;POD1:手术后第1天;POD7:手术后第7天;CRP:C反应蛋白;NLR:中性粒细胞与淋巴细胞比值;SRSS:睡眠状况自评量表得分。*P < 0.05。

3.5. 不同术式中联合预测因子对POCD的预测价值

置换心脏瓣膜的数量对手术进程的影响较大,接受多个心脏瓣膜置换的患者POCD的发生率高于单瓣膜置换的患者(29.55% vs 18.33%),见图2。为分析联合预测因子在不同术式患者中对POCD的预测效果,根据术中置换心脏瓣膜的数量将患者分入不同亚组进行单因素分析,结果显示联合预测因子在单瓣膜置换和多瓣膜置换的患者亚组中与POCD间均存在显著相关性(P < 0.001),见图2。联合预测因子在单瓣膜置换和多瓣膜置换的患者亚组中预测POCD的AUC、最佳截断值、灵敏度、特异度分别为0.9295、0.4041、0.9091、0.9388和0.9082、0.2946、0.7692、0.9355,见图3

4. 讨论

认知功能障碍是心脏瓣膜置换术后的常见并发症,会对患者的预后和生活质量产生不良影响。本研究对208名患者临床资料分析的结果显示,术中尿量、丙泊酚消耗量、POD0 CRP、POD1 CRP、POD0 NLR和POD7 SRSS得分是心脏瓣膜置换术后发生认知功能障碍的独立影响因素。其中,POD0 NLR和POD7 SRSS得分与POCD的关联较为显著,二者每增加1个单位,发生POCD的风险分别增加1.181倍和1.161倍。考虑到术中尿量(OR = 0.998)和丙泊酚消耗量(OR = 1.002)在模型中的实际应用价值较低,使用逻辑回归模型拟合POD0 CRP、POD1 CRP、POD0 NLR和POD7 SRSS得分4个独立指标构建新的联合预测因子,联合预测因子对术后发生POCD的预测效果良好,联合预测因子每增加0.1个单位,术后发生POCD的风险增加2.232倍,且灵敏度和特异度优于单项指标。

Figure 2. Univariate analysis of combined predictive factors and POCD in different surgical procedures

2. 不同术式中联合预测因子与POCD的单因素分析

Figure 3. ROC curves of combined predictive factors for predicting POCD in different surgical procedures

3. 不同术式中联合预测因子预测POCD的ROC曲线

重度心脏瓣膜病患者大多需要接受手术治疗,而心脏手术中由于正中胸骨切开操作引起的创伤和刺激较大、术中行体外循环以及缺血再灌注损伤等原因,术后会发生明显的应激和炎症反应[13]-[15]。这种术后无菌性全身炎症反应在神经认知功能障碍的发生发展过程中起了重要作用[6],使得心脏手术后患者发生谵妄、认知功能障碍、抑郁等神经精神并发症的风险要显著高于非心脏手术患者[8] [16]术中行多心脏瓣膜置换相较于单瓣膜置换,由于操作更复杂且体外循环时间更长,对患者中枢神经系统的影响更加明显,本研究结果显示其术后认知功能障碍的发生率也较高,但在各类型手术患者亚组中联合预测因子对POCD均有较好的预测价值。

既往研究报道术后全身炎症指标升高与发生认知功能障碍间存在关联[17],本研究也发现术后1小时和术后第1天血清中CRP水平高是发生认知功能障碍的独立危险因素,这可能反应了过度的全身炎症反应对中枢神经系统的影响[6]。全身炎症反应通过损害血脑屏障(blood-brain barrier, BBB)的结构和功能,使促炎细胞因子进入中枢神经系统(central nervous system, CNS),引起CNS炎症并激活小胶质细胞,活化的小胶质细胞上调促炎细胞因子的表达进一步加剧神经炎症,最终引起神经元损伤和脑血管周围β-淀粉样蛋白(β-amyloid, Aβ)沉积,导致CNS功能障碍和各种术后神经认知相关并发症[6] [7]同时有研究发现认知功能障碍患者的血清中的中枢神经特异性蛋白(S-100β)水平术后升高明显,这提示全身炎症对中枢神经系统神经胶质细胞的损伤可能也与认知功能障碍的发生相关[18]

本研究发现术后1小时的NLR指数较高也是发生认知功能障碍的独立危险因素。NLR指数偏高可能提示免疫炎症风暴更加严重,是预测术后认知功能障碍的潜在标志物[17],多项非心脏手术的临床研究显示了高NLR与术后认知功能下降之间的相关性[19]-[22]。活化的全身免疫细胞通过激活小胶质细胞,增强活性氧–核苷酸结合寡聚化结构域样受体家族含热蛋白结构域蛋白3 (reactive oxygen species-nucleotide-binding oligomerization domain-like receptor family pyrin domain Containing 3, ROS-NLRP3)等通路表达加剧CNS炎症,进而诱发补体介导的突触丢失和海马神经元焦亡,导致术后认知功能障碍[23] [24]此外,免疫细胞激活后通过抑制磷脂酰肌醇3-激酶/蛋白激酶B/哺乳动物雷帕霉素靶蛋白(Phosphoinositide 3-kinase/Protein Kinase B/Mammalian Target of Rapamycin, PI3K/AKT/mTOR)信号通路,增加了超氧化物歧化酶(superoxide dismutase, SOD)、过氧化脂质(Lipid Peroxidation, LPO)、丙二醛(Malondialdehyde, MDA)等产物的表达,过氧化氢酶(Catalase, CAT)的活性及谷胱甘肽(Glutathione, GSH)等还原性物质的水平下降,这种增强的氧化应激与海马神经元自噬和数量减少相关,对患者的认知功能产生不利影响[25] [26]

本研究还探讨了术后睡眠质量对认知功能的影响,收集了术后第7天的睡眠状况自评量表(SRSS)得分以反映患者术后1周内的睡眠质量,结果发现术后第7天的SRSS得分是发生POCD的独立危险因素。既往有研究报道术前存在的睡眠障碍会对术后神经认知功能的恢复产生不利影响[27],尽管睡眠对认知功能的影响机制尚不明确,但这提示我们应当更加关注患者在围手术期的睡眠情况。

本研究还发现术中尿量少与术后认知功能不良之间存在关联,这可能是因为术中肾血流灌注不足,间接反映了脑血流灌注情况。在心脏手术过程中,大脑和肾脏的功能容易受到血压(BP)波动的影响,术中低血压(intraoperative hypotension, IOH)会减少器官的血流灌注[28]。既往研究发现术中长时间持续性的低血压事件可能会危害术后短期的认知功能[29] [30]这可能是因为持续低血压造成了脑组织灌注不足并引起了神经系统损伤[31],而麻醉过程中使用药物提高平均动脉压以改善器官血流灌注[32],则可能对认知功能的恢复产生积极影响[33]-[35]

本研究也存在一些局限性。首先,我们的研究仅纳入了来自单个临床中心的患者,后续的研究中应扩大招募患者的来源以增强研究结果的代表性。其次,本研究排除了术前存在认知功能障碍的患者(MMSE得分 < 18),这造成了数据缺失,使得研究结果不适用于此类患者手术后认知功能恢复情况的评估。

综上,术中尿量、丙泊酚消耗量、POD0 CRP、POD1 CRP、POD0 NLR和POD7 SRSS得分是心脏瓣膜置换手术后发生认知功能障碍的独立影响因素,基于上述单指标构建的联合预测因子对于心脏瓣膜置换术后的认知功能障碍有良好的预测价值。临床上对术后发生认知功能障碍的高危患者应加强围术期管理,采取减轻全身炎症反应、改善围术期睡眠质量以及预防持续性低血压等措施,以减少认知功能障碍发生,提高患者术后恢复质量。

基金项目

青岛大学附属医院“临床医学+X”基金(QDFY+X202101057)。

参考文献

[1] Evered, L., Silbert, B., Knopman, D.S., Scott, D.A., DeKosky, S.T., Rasmussen, L.S., et al. (2018) Recommendations for the Nomenclature of Cognitive Change Associated with Anaesthesia and Surgery—2018. Anesthesiology, 129, 872-879.
https://doi.org/10.1097/aln.0000000000002334
[2] Glumac, S., Kardum, G., Sodic, L., Supe-Domic, D. and Karanovic, N. (2017) Effects of Dexamethasone on Early Cognitive Decline after Cardiac Surgery. European Journal of Anaesthesiology, 34, 776-784.
https://doi.org/10.1097/eja.0000000000000647
[3] Knipp, S.C., Weimar, C., Schlamann, M., Schweter, S., Wendt, D., Thielmann, M., et al. (2017) Early and Long-Term Cognitive Outcome after Conventional Cardiac Valve Surgery. Interactive CardioVascular and Thoracic Surgery, 24, ivw421.
https://doi.org/10.1093/icvts/ivw421
[4] Nemeth, E., Vig, K., Racz, K., Koritsanszky, K.B., Ronkay, K.I., Hamvas, F.P., et al. (2017) Influence of the Postoperative Inflammatory Response on Cognitive Decline in Elderly Patients Undergoing On-Pump Cardiac Surgery: A Controlled, Prospective Observational Study. BMC Anesthesiology, 17, Article No. 113.
https://doi.org/10.1186/s12871-017-0408-1
[5] Ghaffary, S., Ghaeli, P., Talasaz, A.H., Karimi, A., Noroozian, M., Salehiomran, A., et al. (2017) Effect of Memantine on Post-Operative Cognitive Dysfunction after Cardiac Surgeries: A Randomized Clinical Trial. DARU Journal of Pharmaceutical Sciences, 25, Article No. 24.
https://doi.org/10.1186/s40199-017-0190-0
[6] Subramaniyan, S. and Terrando, N. (2019) Neuroinflammation and Perioperative Neurocognitive Disorders. Anesthesia & Analgesia, 128, 781-788.
https://doi.org/10.1213/ane.0000000000004053
[7] Wang, P., Velagapudi, R., Kong, C., Rodriguiz, R.M., Wetsel, W.C., Yang, T., et al. (2020) Neurovascular and Immune Mechanisms That Regulate Postoperative Delirium Superimposed on Dementia. Alzheimers & Dementia, 16, 734-749.
https://doi.org/10.1002/alz.12064
[8] Bhushan, S., Li, Y., Huang, X., Cheng, H., Gao, K. and Xiao, Z. (2021) Progress of Research in Postoperative Cognitive Dysfunction in Cardiac Surgery Patients: A Review Article. International Journal of Surgery, 95, Article 106163.
https://doi.org/10.1016/j.ijsu.2021.106163
[9] Kahl, U., Callsen, S., Beck, S., Pinnschmidt, H., von Breunig, F., Haese, A., et al. (2021) Health-Related Quality of Life and Self-Reported Cognitive Function in Patients with Delayed Neurocognitive Recovery after Radical Prostatectomy: A Prospective Follow-Up Study. Health and Quality of Life Outcomes, 19, Article No. 64.
https://doi.org/10.1186/s12955-021-01705-z
[10] Glumac, S., Kardum, G. and Karanovic, N. (2019) Postoperative Cognitive Decline after Cardiac Surgery: A Narrative Review of Current Knowledge in 2019. Medical Science Monitor, 25, 3262-3270.
https://doi.org/10.12659/msm.914435
[11] Glumac, S. and Kardum, G. (2021) Is It Time to Redefine Cognitive Dysfunction after Cardiac Surgery? The Importance of Methodological Consistency. Acta Clinica Croatica, 60, 127-130.
https://doi.org/10.20471/acc.2021.60.01.18
[12] Moller, J., Cluitmans, P., Rasmussen, L., Houx, P., Rasmussen, H., Canet, J., et al. (1998) Long-Term Postoperative Cognitive Dysfunction in the Elderly: ISPOCD1 Study. The Lancet, 351, 857-861.
https://doi.org/10.1016/s0140-6736(97)07382-0
[13] Hayashi, Y., Sawa, Y., Nishimura, M., Satoh, H., Ohtake, S. and Matsuda, H. (2003) Avoidance of Full-Sternotomy: Effect on Inflammatory Cytokine Production During Cardiopulmonary Bypass in Rats. Journal of Cardiac Surgery, 18, 390-395.
https://doi.org/10.1046/j.1540-8191.2003.02046.x
[14] Steinberg, B.M., Grossi, E.A., Schwartz, D.S., McLoughlin, D.E., Aguinaga, M., Bizekis, C., et al. (1995) Heparin Bonding of Bypass Circuits Reduces Cytokine Release during Cardiopulmonary Bypass. The Annals of Thoracic Surgery, 60, 525-529.
https://doi.org/10.1016/0003-4975(95)00482-z
[15] Liu, J., Wang, H. and Li, J. (2016) Inflammation and Inflammatory Cells in Myocardial Infarction and Reperfusion Injury: A Double-Edged Sword. Clinical Medicine Insights: Cardiology, 10, CMC.S33164.
https://doi.org/10.4137/cmc.s33164
[16] Fatehi Hassanabad, A., Bahrami, N., Novick, R.J. and Ali, I.S. (2021) Delirium and Depression in Cardiac Surgery: A Comprehensive Review of Risk Factors, Pathophysiology, and Management. Journal of Cardiac Surgery, 36, 2876-2889.
https://doi.org/10.1111/jocs.15610
[17] Seo, C.L., Park, J.Y., Park, J., Kim, H.E., Cho, J., Seok, J., et al. (2021) Neutrophil-Lymphocyte Ratio as a Potential Biomarker for Delirium in the Intensive Care Unit. Frontiers in Psychiatry, 12, Article 729421.
https://doi.org/10.3389/fpsyt.2021.729421
[18] Hollinger, A., Rüst, C.A., Riegger, H., Gysi, B., Tran, F., Brügger, J., et al. (2021) Ketamine vs. Haloperidol for Prevention of Cognitive Dysfunction and Postoperative Delirium: A Phase IV Multicentre Randomised Placebo-Controlled Double-Blind Clinical Trial. Journal of Clinical Anesthesia, 68, Article 110099.
https://doi.org/10.1016/j.jclinane.2020.110099
[19] Zhao, J., Dai, T., Ding, L., Liang, Y., Yuan, W., Jiang, Y., et al. (2023) Correlation between Neutrophil/Lymphocyte Ratio, Platelet/Lymphocyte Ratio and Postoperative Cognitive Dysfunction in Elderly Patients with Esophageal Cancer. Medicine, 102, e33233.
https://doi.org/10.1097/md.0000000000033233
[20] Li, Y., Huang, H. and Le, Y. (2021) Risk Factors and Predictive Value of Perioperative Neurocognitive Disorders in Elderly Patients with Gastrointestinal Tumors. BMC Anesthesiology, 21, Article No. 193.
https://doi.org/10.1186/s12871-021-01405-7
[21] Wang, L., Chen, B., Liu, T., Luo, T., Kang, W. and Liu, W. (2023) Risk Factors for Delayed Neurocognitive Recovery in Elderly Patients Undergoing Thoracic Surgery. BMC Anesthesiology, 23, Article No. 102.
https://doi.org/10.1186/s12871-023-02056-6
[22] Yong, R. and Meng, Y. (2020) Preoperative Neutrophil-Lymphocyte Ratio, an Independent Risk Factor for Postoperative Cognitive Dysfunction in Elderly Patients with Gastric Cancer. Geriatrics & Gerontology International, 20, 927-931.
https://doi.org/10.1111/ggi.14016
[23] Tan, X., Wang, J., Yao, J., Yuan, J., Dai, Y., Sun, M., et al. (2023) Microglia Participate in Postoperative Cognitive Dysfunction by Mediating the Loss of Inhibitory Synapse through the Complement Pathway. Neuroscience Letters, 796, Article 137049.
https://doi.org/10.1016/j.neulet.2023.137049
[24] Zhou, Y., Zhang, Y., Wang, H., Zhang, X., Chen, Y. and Chen, G. (2023) Microglial Pyroptosis in Hippocampus Mediates Sevolfurane-Induced Cognitive Impairment in Aged Mice via ROS-NLRP3 Inflammasome Pathway. International Immunopharmacology, 116, Article 109725.
https://doi.org/10.1016/j.intimp.2023.109725
[25] Chen, M., Han, Y., Que, B., Zhou, R., Gan, J. and Dong, X. (2022) Prophylactic Effects of Sub-Anesthesia Ketamine on Cognitive Decline, Neuroinflammation, and Oxidative Stress in Elderly Mice. American Journal of Alzheimers Disease & Other Dementias, 37.
https://doi.org/10.1177/15333175221141531
[26] Xie, L., Yu, S., Yang, K., Li, C. and Liang, Y. (2017) Hydrogen Sulfide Inhibits Autophagic Neuronal Cell Death by Reducing Oxidative Stress in Spinal Cord Ischemia Reperfusion Injury. Oxidative Medicine and Cellular Longevity, 2017, Article ID: 8640284.
https://doi.org/10.1155/2017/8640284
[27] Li, R., Chen, N., Wang, E., et al. (2021) Correlation between Preoperative Sleep Disorders and Postoperative Delayed Neurocognitive Recovery in Elderly Patients. Journal of Central South University (Medical Sciences), 46, 1251-1259.
[28] Sessler, D.I., Bloomstone, J.A., Aronson, S., Berry, C., Gan, T.J., Kellum, J.A., et al. (2019) Perioperative Quality Initiative Consensus Statement on Intraoperative Blood Pressure, Risk and Outcomes for Elective Surgery. British Journal of Anaesthesia, 122, 563-574.
https://doi.org/10.1016/j.bja.2019.01.013
[29] Duan, W., Zhou, C., Yang, J., Zhang, Y., Li, Z., Ma, D., et al. (2023) A Long Duration of Intraoperative Hypotension Is Associated with Postoperative Delirium Occurrence Following Thoracic and Orthopedic Surgery in Elderly. Journal of Clinical Anesthesia, 88, Article 111125.
https://doi.org/10.1016/j.jclinane.2023.111125
[30] Friedrich, I., Simm, A., Kötting, J., Thölen, F., Fischer, B. and Silber, R. (2009) Cardiac Surgery in the Elderly Patient. Deutsches Ärzteblatt international, 106, 416-422.
https://doi.org/10.3238/arztebl.2009.0416
[31] Vedel, A.G., Rasmussen, L.S., Holmgaard, F. and Nilsson, J.C. (2018) Response by Vedel et al to Letters Regarding Article, “High-Target versus Low-Target Blood Pressure Management during Cardiopulmonary Bypass to Prevent Cerebral Injury in Cardiac Surgery Patients: A Randomized Controlled Trial”. Circulation, 138, 2447-2448.
https://doi.org/10.1161/circulationaha.118.036490
[32] Zhou, N., Liang, X., Gong, J., Li, H., Liu, W., Zhou, S., et al. (2022) S-Ketamine Used during Anesthesia Induction Increases the Perfusion Index and Mean Arterial Pressure after Induction: A Randomized, Double-Blind, Placebo-Controlled Trial. European Journal of Pharmaceutical Sciences, 179, Article 106312.
https://doi.org/10.1016/j.ejps.2022.106312
[33] Gold, J.P., Charlson, M.E., Williams-Russo, P., et al. (1995) Improvement of Outcomes after Coronary Artery Bypass. A Randomized Trial Comparing Intraoperative High versus Low Mean Arterial Pressure. The Journal of Thoracic and Cardiovascular Surgery, 110, 1302-1311.
[34] Charlson, M.E., Peterson, J.C., Krieger, K.H., Hartman, G.S., Hollenberg, J.P., Briggs, W.M., et al. (2007) Improvement of Outcomes after Coronary Artery Bypass II: A Randomized Trial Comparing Intraoperative High versus Customized Mean Arterial Pressure. Journal of Cardiac Surgery, 22, 465-472.
https://doi.org/10.1111/j.1540-8191.2007.00471.x
[35] Larsen, M.H., Draegert, C., Vedel, A.G., Holmgaard, F., Siersma, V., Nilsson, J.C., et al. (2020) Long‐Term Survival and Cognitive Function According to Blood Pressure Management during Cardiac Surgery. A Follow‐Up. Acta Anaesthesiologica Scandinavica, 64, 936-944.
https://doi.org/10.1111/aas.13595